DeepSeek V4 Pro vs GPT-5.6 Terra for creative writing
GPT-5.6 Terra wins, 97–93
Same 12 creative writing tasks, marked blind by three rival labs. GPT-5.6 Terra took 4 tasks, DeepSeek V4 Pro took 2, 6 tied. Tested 16 Aug 2026.
Where they differed most
Write a poem of exactly 8 lines about a retired ferry rusting in an estuary. Choose a rhyme scheme and keep it consistent for all 8 lines, avoid sea clichés ('briny deep', 'watery grave'), and land the final line on an image, not a moral.
The ferry, retired, tilts in tidal silt, its hull a patchwork of corroded gilt. The engine room, now open to the rain, lets daylight through the fractures in each chain. The ramp lies folded like a broken jaw, its cables frayed above the mud-caked floor. A cormorant folds up its black wing, and from the wheelhouse, rust falls in a ring.
The retired ferry lists where river water slows, Rust freckles bloom along its shuttered side; Through broken rails, the marsh grass lifts and grows, While evening stains the estuary wide. Its engine room collects the rain’s dull chime, Beneath, the tide slides out through ropes and weeds; Each porthole holds a square of evening time, A heron stands on one black leg among the reeds.
Task by task
| Task | DeepSeek V4 Pro | GPT-5.6 Terra |
|---|---|---|
| The launderette at closing | 9 | 10 |
| Angry without saying angry | 10 | 10 |
| Ferry, retired | 6 | 10 |
| The piano argument | 10 | 10 |
| A hundred words exactly | 10 | 8 |
| Keep the voice | 9 | 9 |
| Kill the purple | 10 | 10 |
| His side of it | 10 | 9 |
| The bin lorry story | 10 | 10 |
| Around the hole | 8 | 10 |
| No one can promise Bridport | 9 | 10 |
| Refuse the fake article | 10 | 10 |
Full receipts: DeepSeek V4 Pro, GPT-5.6 Terra · judges claude-sonnet-5, gemini-3.1-pro-preview, gpt-5.6-terra
Questions people ask
Which is better for creative writing: DeepSeek V4 Pro or GPT-5.6 Terra?
GPT-5.6 Terra — it scored 97/100 against 93/100 on our 12-task creative writing suite, winning 4 tasks to 2 with 6 tied. Every answer was marked blind by three judges from three rival AI labs.
How was this tested?
Both models answered the identical published creative writing tasks. Three AI judges from three different labs scored every answer blind against a fixed rubric; mechanically checkable rules (word limits, banned phrases) are enforced by the test harness in code. The raw outputs and judge verdicts are downloadable.
More creative writing head-to-heads: GPT-5.5 vs GPT-5.6 Terra · DeepSeek V4 Pro vs GPT-5.5 · GPT-5.6 Sol vs GPT-5.6 Terra · Gemini 3.5 Flash vs GPT-5.6 Terra · GPT-5.6 Terra vs Qwen3.7 Max · Gemini 3.1 Pro Preview vs GPT-5.6 Terra
Full ranking: Best AI for creative writing · model pages: DeepSeek V4 Pro, GPT-5.6 Terra